第12卷第4期2023年4月Vol.12No.4Apr.2023储能科学与技术EnergyStorageScienceandTechnology基于CEEMDAN和ISOA-ELM的锂电池荷电状态预测刘峰，陈海忠（江苏理工学院，江苏常州213000）摘要：锂电池具有能量密度高、输出电压高、无记忆效应等优点，但过充过放电易引发安全事故，精确预测锂电池荷电状态(SOC)让其工作在最佳状态，具有重要现实意义，本文提出了一种基于自适应噪声集成经验模态分解(CEEMDAN)和数据驱动模型组合预测锂离子电池荷电状态的方法，对锂电池原始电流数据进行模态分解，得到多个子序列模态分量，提出一种基于惯性权重与Levy飞行机制的改进海鸥算法(ISOA)，对极限学习机预测模型(ELM)参数进行优化，构建ISOA-ELM锂电池预测模型；训练模型得到锂电池SOC预测结果。实验结果表明，该模型在实际工作中能够更贴合实际SOC，更有利于锂电池工作在最佳状态。关键词：锂离子电池；荷电状态；自适应噪声经验模态分解；极限学习机；海鸥优化算法doi：10.19799/j.cnki.2095-4239.2022.0708中图分类号：TM911文献标志码：A文章编号：2095-4239（2023）04-1244-13Lithium-ionbatterystatepredictionbasedonCEEMDANandISOA-ELMLIUFeng,CHENHaizhong(JiangsuUniversityofTechnology,Changzhou213000,Jiangsu,China)Abstract:Lithiumbatterieshavehighenergydensity,highoutputvoltage,andnomemoryeffect,buttheiroverchargeandover-dischargecancauseaccidents.Inthisregard,accuratepredictionofthestateofcharge(SOC)oflithiumbatteriesservesasthebestcondition.Inthispaper,amethodbasedonanadaptivenoiseintegratedEmpiricalModedecomposition(CEEMDAN)anddata-drivenmodelisproposedtopredictthestateofchargeoflithium-ionbatteries.Theoriginalvoltagedataoflithium-ionbatteriesweremodallydecomposedtoobtainthemodalcomponentsofmultiplesub-sequences.AnimprovedSeagullalgorithm(ISOA)basedontheinertiaweightandtheLevyflightmechanismwasproposed.Theparametersoftheextremelearningmachinepredictionmodelwereoptimized,andISOA-ELMlithiumbatterypredictionmodelwasbuilt.TheSOCpredictionresultsofthelithiumbatterieswereobtainedbytrainingthemodel.TheexperimentalresultsshowthatthemodelcanbetterfittheactualSOCinpracticeandismoreconducivetothelithiumbatteryworkinginthebeststate.Keywords:lithium-ionbatteries;s...
